Senior Research Engineer - Video Synthesis

At UNITH, we're at the forefront of transforming customer journeys with conversational AI. Listed on the ASX, we lead the way in creating lifelike digital humans using synthetic facial movement, voice engineering, and conversational design. With Digital Humans available in over 60 languages and featuring more than 600 voices, we're revolutionizing the conversational AI industry.

The Role:

We are looking for a talented Research Engineer to advance our talking head synthesis technology and drive the next generation of our digital human platform. You'll join our AI/ML team to improve visual quality, reduce latency, and enhance the robustness of our production video synthesis models. This is a hands-on, research-focused role where you'll use your expertise to guide technical decisions and share knowledge that lifts the whole team.

Key Responsibilities:

• Research and implement state-of-the-art techniques in talking head synthesis
• Adapt cutting-edge models (GANs, Diffusion Models, NeRFs, Gaussian Splatting, 3DMM) for production use
• Optimize existing models for improved visual quality while maintaining low latency
• Develop robust solutions that meet real-world deployment requirements
• Bring deep expertise to guide technical decisions and share knowledge with the team
• Work closely with R&D engineers and interns in daily collaboration
• Collaborate with Back-end, Front-end, and Product teams on integration requirements
• Share expertise and insights to elevate the entire team's technical capabilities
• Work with our existing Python/PyTorch/AWS technology stack
• Ensure seamless integration with our conversation services and backend APIs
• Contribute to our real-time video generation pipeline optimization

Requirements:

• 3+ years of experience in computer vision, video generation, or related fields. PhD in a relevant field or equivalent strong industry experience.
• Strong expertise in one or more of:
• Generative Adversarial Networks (GANs)
• Diffusion Models
• Neural Radiance Fields (NeRFs)
• 3D Gaussian Splatting
• 3D Morphable Models (3DMM)
• Proficiency in Python and PyTorch
• Experience with audio-driven avatar generation, face reenactment, or video synthesis

• Proven track record of implementing and improving deep learning models
• Experience reading, understanding, and implementing state-of-the-art research

• Strong communication skills for technical discussions and cross-functional work
• Experience working in collaborative research or engineering environments
• Ability to share knowledge and contribute to team learning

Bonus Points:

• Experience with production deployment of ML models
• Knowledge of AWS cloud services
• Background in real-time computer graphics or video processing
• Publications or contributions to open-source projects in computer vision
• Experience with speech-driven facial animation or multimodal AI systems

What Success will Look Like:

First 6 months:

• Deep understanding of our current video synthesis pipeline and its limitations
• Integration into the team with strong collaborative relationships
• Initial contributions to model improvements and technical insights

First 12 months:

• Delivery of new technology/solution/architecture that demonstrably improves visual quality
• Maintained or improved latency performance compared to current production models
• Established as a key technical contributor and knowledge resource for the team

What We Offer:

• Competitive salary: €50,000 - €70,000 based on experience
• Flexible work: Hybrid model in Barcelona or full remote options
• Make it real: See your research powering digital human experiences worldwide.
• Expert collaboration: Work alongside experienced engineers in a knowledge-sharing environment
• Impact: Your work directly powers digital human experiences for global enterprises

• Additional perks include a laptop, a great office in the center of Barcelona, a lunch compensation for when in the office, private health insurance with Alan, travel allowance, and a generous ClassPass discount.

To apply, please submit your CV and a short motivation explaining why you’re excited to join UNITH via the ‘easy apply’ button.

For a standout application, you can also reach out directly to [email protected], feel free to be as creative as you’d like 🤖

Our recruitment process looks like this:

1. (Online) Intro meeting with Anniek
2. (Online) meeting with Boris (Product Lead) and Marcel (Head of AI and Platform Engineering)
3. (Online) Meeting with the R&D team
4. Small assignment (not too time-consuming, just a way to see how you think and approach problems)
5. Reference check
   

At UNITH, we believe diversity drives innovation. Our mission to transform customer experiences relies on a team with a wide range of perspectives, experiences, and backgrounds. We are committed to creating an inclusive workplace where everyone feels valued and empowered to thrive.